CN112801529B - Financial data analysis method and device, electronic equipment and medium - Google Patents
Financial data analysis method and device, electronic equipment and medium Download PDFInfo
- Publication number
- CN112801529B CN112801529B CN202110164468.4A CN202110164468A CN112801529B CN 112801529 B CN112801529 B CN 112801529B CN 202110164468 A CN202110164468 A CN 202110164468A CN 112801529 B CN112801529 B CN 112801529B
- Authority
- CN
- China
- Prior art keywords
- threshold
- financial
- index
- analyzed
- value
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 43
- 238000007405 data analysis Methods 0.000 title claims abstract description 34
- 230000007246 mechanism Effects 0.000 claims abstract description 42
- 238000007477 logistic regression Methods 0.000 claims abstract description 35
- 230000002159 abnormal effect Effects 0.000 claims description 31
- 230000008859 change Effects 0.000 claims description 30
- 230000007306 turnover Effects 0.000 claims description 23
- 230000001419 dependent effect Effects 0.000 claims description 20
- 238000004364 calculation method Methods 0.000 claims description 16
- 230000008520 organization Effects 0.000 claims description 13
- 238000012216 screening Methods 0.000 claims description 13
- 238000007781 pre-processing Methods 0.000 claims description 11
- 238000012545 processing Methods 0.000 claims description 10
- 230000008901 benefit Effects 0.000 claims description 9
- 238000004590 computer program Methods 0.000 claims description 8
- 238000003860 storage Methods 0.000 claims description 8
- 238000012217 deletion Methods 0.000 claims description 6
- 230000037430 deletion Effects 0.000 claims description 6
- 239000000523 sample Substances 0.000 claims 19
- 239000000758 substrate Substances 0.000 claims 4
- 238000004458 analytical method Methods 0.000 description 7
- 238000004891 communication Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 3
- 230000007774 longterm Effects 0.000 description 3
- 230000005856 abnormality Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000002360 preparation method Methods 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000003066 decision tree Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 230000006735 deficit Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000010422 painting Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000011425 standardization method Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012706 support-vector machine Methods 0.000 description 1
- 230000004083 survival effect Effects 0.000 description 1
- 208000024891 symptom Diseases 0.000 description 1
- 238000012549 training Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0635—Risk analysis of enterprise or organisation activities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/12—Accounting
- G06Q40/125—Finance or payroll
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Finance (AREA)
- Accounting & Taxation (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Entrepreneurship & Innovation (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Development Economics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Educational Administration (AREA)
- Tourism & Hospitality (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Game Theory and Decision Science (AREA)
- Technology Law (AREA)
- Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
Abstract
The invention provides a financial data analysis method and device, electronic equipment and medium, wherein the method comprises the following steps: inputting the financial data to be analyzed of the enterprise to be analyzed into a logistic regression model to obtain a current financial false making probability value; inputting Themis the financial data to be analyzed into a financial risk model to obtain a current risk score; determining the risk grade of the mechanism to be analyzed corresponding to the financial data to be analyzed by utilizing the current financial counterfeiting probability and the current risk score, analyzing the financial data to be analyzed by combining a logistic regression model and a Themis financial risk model to obtain a counterfeiting probability value and a risk score, and jointly applying the models to obtain a final model result and a risk grade division standard after the two models are jointly applied, so that the accuracy is effectively improved, a professional and efficient financial data analysis result can be formed, and risks are found and avoided.
Description
Technical Field
The present invention relates to the field of data processing technologies, and in particular, to a method and apparatus for analyzing financial data, an electronic device, and a medium.
Background
At present, most of the fake-making analysis of financial data of the institutions to be analyzed is mainly based on financial experts and artificial experience of accounting to judge the financial report subject abnormality and fake suspicion of the institutions to be analyzed, and under the condition of large data volume, the efficiency and quality are difficult to guarantee, and the method is difficult to quickly find the financial abnormality of the institutions to be analyzed in the market. And Themis the financial risk model is a classical financial risk model, but for actual financial falsification data, the accuracy rate of the model has a space to be improved.
Therefore, how to provide a financial data analysis scheme, which can form professional and efficient financial data analysis results, and find and avoid risks are technical problems to be solved urgently by those skilled in the art.
Disclosure of Invention
The invention provides a financial data analysis method and device, electronic equipment and medium, which can form professional and efficient financial data analysis results, and discover and avoid risks.
In a first aspect, the present invention provides a method of financial data analysis comprising:
Inputting the financial data to be analyzed of the mechanism to be analyzed into a logistic regression model to obtain a current financial false making probability value;
inputting Themis the financial data to be analyzed into a financial risk model to obtain a current risk score;
and determining the risk grade of the institution to be analyzed corresponding to the financial data to be analyzed by using the current financial falsification probability and the current risk score.
Further, the logistic regression model is obtained by:
Acquiring sample data; the sample data comprise good samples and bad samples, wherein the good samples are financial and newspaper data of the same industry with the same preset time period of the mechanism to be analyzed; the bad sample is fake financial data of the same industry as the organization to be analyzed;
Preprocessing the sample data;
screening independent variables and dependent variables from the preprocessed sample data;
Determining the correlation coefficient of the independent variable and the dependent variable, and screening out the independent variable with the correlation exceeding a preset value; clustering the independent variables to obtain a new independent variable set;
And performing logistic regression on the new independent variable set and the dependent variable to obtain an effective prediction independent variable set.
Further, the preprocessing the sample data includes:
deleting data with the number of data deletions exceeding a preset number in the sample data;
Replacing an abnormal value in the sample data;
And (5) carrying out standardization processing on the sample data.
Further, the inputting Themis the financial data to be analyzed into a financial risk model, and obtaining the current risk score includes:
determining an index result of a core index based on the financial data to be analyzed;
Obtaining a scoring value of the index result based on the index result, the corresponding index threshold and the scoring rule;
adding the scoring values of all indexes to obtain a preliminary risk score of the mechanism to be analyzed;
And adjusting the preliminary risk score by using an adjustment item to obtain the current risk score of the mechanism to be analyzed.
Further, the core metrics include at least one of: the balance period level index, the balance period change amount index, the financial debt state level index, the financial debt state change amount index, the physical asset turnover change amount, the investment benefit level, the intangible asset efficiency change amount, the frequent balance ratio, the abnormal value coefficient, the payment remaining force coefficient, the asset coefficient, the cost system and other indexes, the mobile liability occupation ratio and the monetary funds occupation proportion.
Further, the index threshold includes: a first threshold, a second threshold, a third threshold, and a fourth threshold;
The first threshold is a threshold which adopts the first 10% quantile value as an extreme abnormal point for the whole index;
The second threshold is a threshold which adopts the last 10% quantile value as an extreme abnormal point for the whole index;
The third threshold value is determined by the following steps: calculating the mean value of each dimension as a median result after calculating the first 10% of data, calculating standard deviation, respectively calculating standard deviation multiples of a first threshold point, and determining the standard deviation multiples as a third threshold according to the standard deviation multiples which are half of the standard deviation multiples;
the fourth threshold value is determined by the following steps: and calculating the mean value of each dimension as a median result after 10% of data are calculated, calculating standard deviation, respectively calculating standard deviation multiples of a second threshold, and determining a fourth threshold according to the fact that one half of the standard deviation multiples are used as the standard deviation multiples of the fourth threshold.
Further, the scoring rule is: setting a respective score calculation rule formula for each core index, and obtaining the lowest score when the index result exceeds a first threshold value and a second threshold value of the index;
When the index result is between the first threshold value and the third threshold value or between the second threshold value and the fourth threshold value, obtaining a corresponding score by adopting linear calculation according to the index result, the threshold value and the score standard;
And when the index result is between the third threshold value and the fourth threshold value, obtaining a unified standard score.
Further, the determining the risk level of the institution to be analyzed corresponding to the financial data to be analyzed by using the current financial falsification probability and the current risk score includes:
Acquiring a corresponding relation between a risk level and a first range value of a financial falsification probability value and between the risk level and a second range value of a risk score;
And determining the risk level of the institution to be analyzed based on a first matching relation between the current financial falsification probability value and the first range value and a second matching relation between the current risk score and the second range value.
In a second aspect, the present invention provides a financial data analysis apparatus comprising:
the probability acquisition module is used for inputting the financial data to be analyzed of the mechanism to be analyzed into the logistic regression model to obtain the current financial false-making probability value;
The score acquisition module is used for inputting Themis the financial data to be analyzed into a financial risk model to obtain a current risk score;
and the grade determining module is used for determining the risk grade of the institution to be analyzed corresponding to the financial data to be analyzed by utilizing the current financial false making probability and the current risk score.
Further, the logistic regression model is obtained by:
Acquiring sample data; the sample data comprise good samples and bad samples, wherein the good samples are financial and newspaper data of the same industry with the same preset time period of the mechanism to be analyzed; the bad sample is fake financial data of the same industry as the organization to be analyzed;
Preprocessing the sample data;
screening independent variables and dependent variables from the preprocessed sample data;
Determining the correlation coefficient of the independent variable and the dependent variable, and screening out the independent variable with the correlation exceeding a preset value; clustering the independent variables to obtain a new independent variable set;
And performing logistic regression on the new independent variable set and the dependent variable to obtain an effective prediction independent variable set.
Further, the preprocessing the sample data includes:
deleting data with the number of data deletions exceeding a preset number in the sample data;
Replacing an abnormal value in the sample data;
And (5) carrying out standardization processing on the sample data.
Further, the score acquisition module includes:
An index determining unit for determining an index result of a core index based on the financial data to be analyzed;
The scoring obtaining unit is used for obtaining a scoring value of the index result based on the index result, the corresponding index threshold value and the scoring rule;
the preliminary obtaining unit is used for adding the scoring values of all indexes to obtain a preliminary risk score of the mechanism to be analyzed;
And the grading adjustment unit is used for adjusting the preliminary risk grading by using the adjustment items to obtain the current risk grading of the mechanism to be analyzed.
Further, the core metrics include at least one of: the balance period level index, the balance period change amount index, the financial debt state level index, the financial debt state change amount index, the physical asset turnover change amount, the investment benefit level, the intangible asset efficiency change amount, the frequent balance ratio, the abnormal value coefficient, the payment remaining force coefficient, the asset coefficient, the cost system and other indexes, the mobile liability occupation ratio and the monetary funds occupation proportion.
Further, the index threshold includes: a first threshold, a second threshold, a third threshold, and a fourth threshold;
The first threshold is a threshold which adopts the first 10% quantile value as an extreme abnormal point for the whole index;
The second threshold is a threshold which adopts the last 10% quantile value as an extreme abnormal point for the whole index;
The third threshold value is determined by the following steps: calculating the mean value of each dimension as a median result after calculating the first 10% of data, calculating standard deviation, respectively calculating standard deviation multiples of a first threshold point, and determining the standard deviation multiples as a third threshold according to the standard deviation multiples which are half of the standard deviation multiples;
the fourth threshold value is determined by the following steps: and calculating the mean value of each dimension as a median result after 10% of data are calculated, calculating standard deviation, respectively calculating standard deviation multiples of a second threshold, and determining a fourth threshold according to the fact that one half of the standard deviation multiples are used as the standard deviation multiples of the fourth threshold.
Further, the scoring rule is: setting a respective score calculation rule formula for each core index, and obtaining the lowest score when the index result exceeds a first threshold value and a second threshold value of the index;
When the index result is between the first threshold value and the third threshold value or between the second threshold value and the fourth threshold value, obtaining a corresponding score by adopting linear calculation according to the index result, the threshold value and the score standard;
And when the index result is between the third threshold value and the fourth threshold value, obtaining a unified standard score.
Further, the rank determination module includes:
the relationship obtaining unit is used for obtaining the corresponding relationship between the risk level and the first range value of the financial falsification probability value and the corresponding relationship between the risk level and the second range value of the risk score;
And the level matching unit is used for determining the risk level of the mechanism to be analyzed based on the first matching relation between the current financial falsification probability value and the first range value and the second matching relation between the current risk score and the second range value.
In a third aspect, the invention also provides an electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the steps of any of the financial data analysis methods described above when the program is executed.
In a fourth aspect, the present invention also provides a non-transitory computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements the steps of a financial data analysis method as described in any of the above.
According to the financial data analysis method and device, the electronic equipment and the medium, the financial data to be analyzed is analyzed by combining the logistic regression model and the Themis financial risk model, the fake probability value and the risk score are obtained, the models are jointly applied, so that a final model result and a risk grade division standard are obtained after the two models are jointly applied, the accuracy is effectively improved, a professional and efficient financial data analysis result can be formed, and risks are found and avoided.
Drawings
In order to more clearly illustrate the invention or the technical solutions of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are some embodiments of the invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of a method for analyzing financial data according to an embodiment of the present invention;
FIG. 2 is a second flowchart of a method for analyzing financial data according to an embodiment of the present invention;
FIG. 3 is a third flowchart of a method for analyzing financial data according to an embodiment of the present invention;
FIG. 4 is a schematic diagram of a composition structure of a financial data analysis device according to an embodiment of the present invention;
fig. 5 is a schematic structural diagram of an electronic device provided by the present invention;
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the present invention more apparent, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings, and it is apparent that the described embodiments are some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
A financial data analysis method of the present invention is described below in connection with fig. 1-3.
FIG. 1 is a flow chart of a method for analyzing financial data according to an embodiment of the present invention; FIG. 2 is a second flowchart of a method for analyzing financial data according to an embodiment of the present invention; FIG. 3 is a third flowchart of a method for analyzing financial data according to an embodiment of the present invention.
In a specific implementation manner of the present invention, an embodiment of the present invention provides a financial data analysis method, including:
Step 110: inputting the financial data to be analyzed of the mechanism to be analyzed into a logistic regression model to obtain a current financial false making probability value;
Specifically, in the embodiment of the invention, firstly, the financial data to be analyzed of the organization to be analyzed needs to be acquired, generally, if the organization is a marketing company, the financial report of the year and the quarter needs to be issued at a specified time, and of course, the organization to be analyzed can also submit the financial data to be analyzed by itself.
In the embodiment of the application, the logic regression model is adopted to analyze the fake probability of the financial data to be analyzed. In practice, institutions perform financial counterfeiting in three general types, and if institutions increase profits virtually, the institutions must transfer to the liability statement through unassigned profit projects, resulting in either a virtual increase or a virtual decrease in liabilities. The first way is: the mode of increasing the receivables in a virtual way is simple and rough, is easy to identify, and is a primary fake making mode. The receivables will not generate operational cash flow, so whether the cash flow net amount/net profit value is abnormal or not can be found by comparing the receivables and selling cash/business income received by the commodity. In connection with institutional operations, if not properly interpreted, there is often a potential for counterfeiting. The second mode is as follows: in addition, the money is easily checked by many people, and is not easy to be counterfeited, but many cases tell us that the money is counterfeited frequently by means of private bank seals, counterfeit bank running water, receipt, false reply, false online banking, fake banking sites and the like. The new technology of Kangde in the present year is to push money funds fake means to climax by adopting a modern means of 'funds collection'. Third type of mode: the money is virtually increased and then converted into other assets, the counterfeiter makes up complete transactions and finishes the money making, no problem is found in the operation cash flow, but the money is quickly converted into other assets, and the money is slowly digested through asset value reduction means such as bad account preparation, price drop preparation, depreciation and the like.
The embodiment of the invention adopts a logistic regression model, can find out the internal linear relation of various parameters in the financial data of the mechanism, and judges the degree of deviation of the financial data to be analyzed from a normal value to judge the fake probability of the financial data.
Step 120: inputting Themis the financial data to be analyzed into a financial risk model to obtain a current risk score;
in the embodiment of the application, the Themis financial risk model can be used for carrying out risk scoring on the financial data to be analyzed, and the Themis financial risk model is used for carrying out the financial risk scoring, so that the enterprise operation safety is higher and the enterprise operation risk is smaller. Specifically, the index expansion can be performed on the basis of the core index thereof, and besides indexes including the level index of the period of the liability and liability, the index of the change amount of the period of the liability and liability, the level index of the condition of the liability and liability, the index of the change amount of the condition of the liability and liability, the physical asset turnover change amount, the investment benefit level, the intangible asset efficiency change amount, the frequent balance ratio, the outlier coefficient, the payment remainder coefficient, the asset coefficient, the cost system and the like, the mobile liability ratio and the monetary funds occupation proportion are increased to be used as adjustment items, and the financial data to be analyzed are scored on each index to obtain the current risk score.
Step 130: and determining the risk grade of the institution to be analyzed corresponding to the financial data to be analyzed by using the current financial falsification probability and the current risk score.
The financial data to be analyzed is analyzed by combining the logistic regression model and the Themis financial risk model, a fake probability value and a risk score are obtained, the models are combined to obtain a final model result and a risk grade division standard after the two models are combined to be applied, the accuracy is effectively improved, a professional and efficient financial data analysis result can be formed, and risks are found and avoided.
Further, in one embodiment the logistic regression model is obtained by: acquiring sample data; the sample data comprise good samples and bad samples, wherein the good samples are financial and newspaper data of the same industry with the same preset time period of the mechanism to be analyzed; the bad sample is fake financial data of the same industry as the organization to be analyzed; preprocessing the sample data; screening independent variables and dependent variables from the preprocessed sample data; determining the correlation coefficient of the independent variable and the dependent variable, and screening out the independent variable with the correlation exceeding a preset value; clustering the independent variables to obtain a new independent variable set; and performing logistic regression on the new independent variable set and the dependent variable to obtain an effective prediction independent variable set.
Specifically, in order to preprocess the sample data, deleting data with the number of data deletion exceeding a preset number in the sample data; replacing an abnormal value in the sample data; and (5) carrying out standardization processing on the sample data by using a maximum and minimum method. More specifically, missing value serious data is deleted: if only the number of non-missing variables is reserved and is more than 0.5, the number of independent variables enters model training and the like. Outlier and missing value processing: if the data with the bit number smaller than 1 and the bit number larger than 99 is regarded as the abnormal value, a replacement principle is formulated, and the abnormal value is replaced correspondingly. Data standardized disassembly processing: the data is standardized by adopting a maximum and minimum method, and other data standardization methods can be adopted; when the maximum and minimum methods are standardized, the standard value can be calculated by adopting the following formula: x= (X-x.min)/(x.max-x.min); where x is the data to be normalized, x.min is the minimum value in x, and x.max is the maximum value in x.
Sample data for a logistic regression model can be divided into two parts: official published information such as financial falsification company, falsification report period and the like; expert judges the marked high probability faking company and the corresponding report; model quality sample determination: good samples: only the financial data of the marketing company in the same year and the same industry as the fake company is selected as a good sample data set; bad samples: for data that is counterfeited for a plurality of years in succession by the same marketing company, only the first counterfeited year is reserved to enter the bad sample data set.
In order to screen the independent variables, firstly, calculating the correlation coefficient between the independent variables and the dependent variables, and screening out the independent variables with strong correlation; and removing the collinearity by a clustering method to obtain a new independent variable set. Performing logistic regression on the new independent variable set and the dependent variable to obtain a group of independent variable sets with the most effective prediction effect; and the independent variable set is used for a decision tree and a support vector machine, the two model results are compared with a logistic regression model, and the logistic regression model effect is obtained to be relatively optimal according to the recall rate and the accuracy rate.
In yet another embodiment of the present invention, in order to input Themis the financial data to be analyzed into a financial risk model, the following steps may be specifically performed to obtain a current risk score:
Step 210: determining an index result of a core index based on the financial data to be analyzed;
Specifically, the core index includes at least one of: the balance period level index, the balance period change amount index, the financial debt state level index, the financial debt state change amount index, the physical asset turnover change amount, the investment benefit level, the intangible asset efficiency change amount, the frequent balance ratio, the abnormal value coefficient, the payment remaining force coefficient, the asset coefficient, the cost system and other indexes, the mobile liability occupation ratio and the monetary funds occupation proportion.
Step 220: obtaining a scoring value of the index result based on the index result, the corresponding index threshold and the scoring rule;
Step 230: adding the scoring values of all indexes to obtain a preliminary risk score of the mechanism to be analyzed;
Step 240: and adjusting the preliminary risk score by using an adjustment item to obtain the current risk score of the mechanism to be analyzed.
The meaning of some core metrics is described below:
(1) The charge purchase turnover is scored according to a matrix table of the grade and the variation, wherein the grade is obtained by subtracting the calculated formula of the sales performance divided by the month after the accounts receivable is deducted from the accounts receivable, and the difference between the charge purchase turnover and the turnover before the 2-phase is taken as the 'variation' of the turnover. The turnover period of the survival mechanism is stable, and the switching mechanism can change between positive and negative values greatly. In particular, if poor creditor or receivables are stably recovered and stopped, the index of the credited debt period changes greatly in the positive direction and a break-off occurs, or if a break-off occurs in which an attempt to evade funds with difficulty by increasing payable bill settlement occurs, the index of the period changes greatly in the positive direction. This condition, in fact, is manifested as a symptom at the end of the business.
(2) The financial liability sales ratio compares the conditions that short-term borrowing, long-term liabilities due in one year, long-term borrowing, bonds payable and the like can be regarded as borrowing with the monthly marketing amount according to the characteristics of industries, and judges whether the borrowing quantity of the institutions is reasonable compared with the same industry. Such as: the credit dependency between institutions such as accounts payable in retail industry and architecture industry is large, while the borrowing scale is not so large. However, on the contrary, it means that the guarantee of borrowing money to a bank is weak, and even if the bank is frustrated with a small frustration, the bank is not easy to support, and the burst of fund collection is easy to be exposed, which is a common situation. In practice, the retail, wholesale and construction industries are more fragile than the manufacturing industries that are misclassified to banks with physically fixed assets as warranties.
(3) The financial liability is not sound, the variation of the borrowing scale is evaluated through the increase and decrease of the borrowing quantity in the earlier stage, and the variation is compared with the monthly sales to judge whether the borrowing is reasonable in business use or not, and whether the borrowing is for making up the defect or paying the liability in the earlier stage or not, and the risk degree exists for the current operation of the institutions. The borrowing is usually performed for normal business operations, which brings benefits in a short period of time, and if the turnover funds are added for equipment investment, inventory investment, and unused sales, or the borrowing is added for filling the deficiency, the risk of borrowing increases.
(4) Physical asset turnover, 1) inventory turnover, when an organization is operating normally, inventory turnover is maintained at a relatively stable level, but if there is a shortage of funds, funds are raised by processing the inventory, or if there is a crisis in the operation, inventory will be sold, different industries are distinguished, and inventory turnover will vary abnormally. 2) Fixed asset turnover variability, which is the variation between the investment of the fixed asset in the institution and the sales of the institution, is analyzed, thereby revealing the efficiency of the investment of the fixed asset in the institution. The hysteresis of the institution investment in the period of the fixed asset and the investment over a period of time in sales is also fully taken into account in the analysis process.
(5) And the investment asset efficiency is used for judging the profit condition of the investment asset of the institution according to the comparison of the investment asset and the investment profit of the institution in a certain period.
(6) And judging possible painting or profit conditions of the mechanism intangible asset according to the change and profit conditions of the mechanism intangible asset.
(7) The frequent balance ratio, the ratio of frequent activity income to expense of the calculating mechanism is calculated through the combination of the asset liability statement and the profit statement, and the comprehensive consideration of the frequent balance ratio in the last three years is used for scoring. From the correlation between the frequent balance and the reclosing mechanism, it can be derived that: the statistics result that the probability of closing is higher when the count rate is even 80% of the time when the count rate is always within 95% to 100% of the time when the count rate is 2 to 3.
(8) And comparing the sales, inventory turnover and turnover conditions of liabilities by using the anomaly coefficient, calculating the unbalance amount between the sales and inventory turnover conditions, calculating the anomaly degree of the anomaly value relative to the sales income, and judging the anomaly condition in the credit transaction of the institution.
(9) Themis pay the remainder factor, judging the ability of a facility to resist risk from a dynamic and static perspective by the turnover of the asset, the return of the own capital with real payment capabilities. Such as: if the annual sales scale for an organization is 1000, then the organization's own capital would be offset by the occurrence of a 5% deficit or poor bond, with an own capital of 50 (net capital to total capital ratio of 50%). In a general financial data analysis method, security is evaluated in terms of the ratio of net value to gross capital, profitability is evaluated in terms of sales versus regular profits, and balance between analysis of liability and analysis of profit margins is maintained. In the "Themis financial risk model", there is a feature of matching both the liability statement and the profit statement together for analysis. In reality, if an institution has poor creditability, the amount of payment remaining for the institution's final safeguarding means must be based on empirical criteria that are minimally met. From the example of the switching mechanism, the payment balance as a safeguard must be about 2 times the monthly marketing performance. In summary, this index has the lowest allowable limit that reflects poor creditor or whitewashing.
10. Cost system, by analyzing the relationship between income and cost, the short-term profitability and competitiveness of an organization are evaluated.
11. Asset coefficients, long term returns from the asset turnover and return situation judgment institution.
Further, in one embodiment of the present invention, the index threshold includes: a first threshold, a second threshold, a third threshold, and a fourth threshold;
The first threshold is a threshold which adopts the first 10% quantile value as an extreme abnormal point for the whole index;
The second threshold is a threshold which adopts the last 10% quantile value as an extreme abnormal point for the whole index;
The third threshold value is determined by the following steps: calculating the mean value of each dimension as a median result after calculating the first 10% of data, calculating standard deviation, respectively calculating standard deviation multiples of a first threshold point, and determining the standard deviation multiples as a third threshold according to the standard deviation multiples which are half of the standard deviation multiples;
the fourth threshold value is determined by the following steps: and calculating the mean value of each dimension as a median result after 10% of data are calculated, calculating standard deviation, respectively calculating standard deviation multiples of a second threshold, and determining a fourth threshold according to the fact that one half of the standard deviation multiples are used as the standard deviation multiples of the fourth threshold.
The scoring rule is as follows: setting a respective score calculation rule formula for each core index, and obtaining the lowest score when the index result exceeds a first threshold value and a second threshold value of the index; when the index result is between the first threshold and the third threshold or between the second threshold and the fourth threshold, corresponding scores are obtained by adopting linear calculation according to the index result, the threshold and the score standard, and when the index result is between the third threshold and the fourth threshold, unified standard scores are obtained.
That is, for the index threshold calculation: better worse threshold: the first 10% and the last 10% of the quantile values are adopted as threshold levels of extreme abnormal points for the whole indexes; good and bad thresholds: calculating the mean value of each dimension as a median result after calculating the data of the front 10% and the rear 10%, calculating standard deviation, calculating the standard deviation multiple of a better threshold point and a worse threshold point respectively, and calculating the threshold of the two points according to the half of the standard deviation multiple as the standard deviation multiple of the better threshold point and the worse threshold point. For the index scoring rule: setting a specific score calculation rule formula for each index, wherein the basic principle of the formula is that the lowest score is obtained when the basic principle of the formula exceeds the respective better and worse thresholds of the indexes; when the value is between the good threshold value and between the poor threshold value and the poor threshold value, corresponding scores are obtained by adopting linear calculation according to the index value, the threshold value and the score standard; when between good and bad, a unified standard score is obtained. The scores of the indexes are divided into 3 three grades, namely standard grade, good grade and lowest grade, the scores of the gears of each index are not completely the same, the scores of the gears of each index are obtained through comprehensive evaluation according to historical data statistics and expert experience, and each index is 0 to 10 grades. Each index obtains corresponding financial reporting period data according to a formula, and calculates the corresponding financial reporting period data respectively, wherein the indexes of the result can be calculated by requiring multi-period data, and when the indexes cannot be calculated due to data missing, the index calculation result is regarded as an abnormal value and does not enter the model. When the calculated result in the model is abnormal more than a certain proportion (more than 50% according to statistics and expert experience), the corresponding subject model result is regarded as abnormal and has no value.
The model score of the mechanism consists of two parts: the indexes are respectively scored and added to obtain the comprehensive model score of the mechanism. The score adjustment value is mainly obtained by carrying out joint threshold management on indexes with two indexes of horizontal quantity and fluctuation quantity. Under Themis financial risk model, about 70% of accuracy is obtained for model sample data of specific false financial report, but not specific bankruptcy or ST institution, and the model is used as a classical quantitative model, so that the accuracy is better, but the accuracy is improved compared with 80% of accuracy judged by manual analysis in the industry.
On the basis of the above embodiment, in this embodiment, determining the risk level of the institution to be analyzed corresponding to the financial data to be analyzed by using the current financial falsification probability and the current risk score may specifically include the following steps:
Step 310: acquiring a corresponding relation between a risk level and a first range value of a financial falsification probability value and between the risk level and a second range value of a risk score;
Step 320: and determining the risk level of the institution to be analyzed based on a first matching relation between the current financial falsification probability value and the first range value and a second matching relation between the current risk score and the second range value.
The logistic regression model and Themis financial risk model are combined, and as the special financial data analysis attribute of the Themis financial risk model is adopted, the overall model takes the Themis financial risk model as a main shaft, the results of the logistic regression model are further combined and mapped according to the division basis of the scores and the risk grades, the final model results and the risk grade division standard are obtained after the two models are combined and applied, the following false making probability is the false making probability value obtained by the logistic regression model, the current risk score obtained by the Themis financial risk model is obtained, and the risks are sequentially from high to low:
Risk class III: probability of falsification > =0.9 and score < =40;
Risk class II: 1:probability of fraud > = 0.9 and 40< score < = 60;2:0.7< = false creation probability <0.9 and score < = 60;
Risk class I1:0.6 < = false creation probability <0.7 and score < = 60;2: probability of falsification > = 0.70;
Attention: 1:0.4< = probability of counterfeiting <0.6;2: probability of falsification > = 0.6 and score >60;
Security 0.2< = false creation probability <0.4 and score <60; the probability of false creation is <0.2.
Under the above risk level classification, 80% or more of the counterfeited samples are classified into the risk level I and above, and the proportion of classified into the risk levels III and II reaches 70%. On the whole model result, the accuracy is effectively improved, and the accuracy is used as a pure quantitative model and approaches to the judgment of the manual experience of an industry expert.
Of course, other risk levels may be set, corresponding to the sum score of different faking probabilities, which is only illustrative and not limiting.
According to the financial data analysis method provided by the embodiment of the invention, the financial data to be analyzed is analyzed by combining the logistic regression model and the Themis financial risk model to obtain the fake probability value and the risk score, the models are jointly applied to obtain the final model result and the risk grade division standard after the two models are jointly applied, the accuracy is effectively improved, the professional and efficient financial data analysis result can be formed, and the risk is found and avoided.
The present invention will be described below with reference to a financial data analysis apparatus, which is described below, and a financial data analysis method described above, which are referred to in correspondence with each other.
Referring to fig. 4, fig. 4 is a schematic diagram illustrating a composition structure of a financial data analysis device according to an embodiment of the invention.
In yet another embodiment of the present invention, an embodiment of the present invention provides a financial data analysis apparatus 400, comprising:
The probability acquisition module 410 is configured to input financial data to be analyzed of the mechanism to be analyzed into a logistic regression model to obtain a current financial modeling probability value;
the score obtaining module 420 is configured to input Themis the financial data to be analyzed into a financial risk model to obtain a current risk score;
the level determining module 430 is configured to determine a risk level of the institution to be analyzed corresponding to the financial data to be analyzed by using the current financial falsification probability and the current risk score.
Further, the logistic regression model is obtained by:
Acquiring sample data; the sample data comprise good samples and bad samples, wherein the good samples are financial and newspaper data of the same industry with the same preset time period of the mechanism to be analyzed; the bad sample is fake financial data of the same industry as the organization to be analyzed;
Preprocessing the sample data;
screening independent variables and dependent variables from the preprocessed sample data;
Determining the correlation coefficient of the independent variable and the dependent variable, and screening out the independent variable with the correlation exceeding a preset value; clustering the independent variables to obtain a new independent variable set;
And performing logistic regression on the new independent variable set and the dependent variable to obtain an effective prediction independent variable set.
Further, the preprocessing the sample data includes:
deleting data with the number of data deletions exceeding a preset number in the sample data;
Replacing an abnormal value in the sample data;
And (5) carrying out standardization processing on the sample data.
Further, the score acquisition module includes:
An index determining unit for determining an index result of a core index based on the financial data to be analyzed;
The scoring obtaining unit is used for obtaining a scoring value of the index result based on the index result, the corresponding index threshold value and the scoring rule;
the preliminary obtaining unit is used for adding the scoring values of all indexes to obtain a preliminary risk score of the mechanism to be analyzed;
And the grading adjustment unit is used for adjusting the preliminary risk grading by using the adjustment items to obtain the current risk grading of the mechanism to be analyzed.
Further, the core metrics include at least one of: the balance period level index, the balance period change amount index, the financial debt state level index, the financial debt state change amount index, the physical asset turnover change amount, the investment benefit level, the intangible asset efficiency change amount, the frequent balance ratio, the abnormal value coefficient, the payment remaining force coefficient, the asset coefficient, the cost system and other indexes, the mobile liability occupation ratio and the monetary funds occupation proportion.
Further, the index threshold includes: a first threshold, a second threshold, a third threshold, and a fourth threshold;
The first threshold is a threshold which adopts the first 10% quantile value as an extreme abnormal point for the whole index;
The second threshold is a threshold which adopts the last 10% quantile value as an extreme abnormal point for the whole index;
The third threshold value is determined by the following steps: calculating the mean value of each dimension as a median result after calculating the first 10% of data, calculating standard deviation, respectively calculating standard deviation multiples of a first threshold point, and determining the standard deviation multiples as a third threshold according to the standard deviation multiples which are half of the standard deviation multiples;
the fourth threshold value is determined by the following steps: and calculating the mean value of each dimension as a median result after 10% of data are calculated, calculating standard deviation, respectively calculating standard deviation multiples of a second threshold, and determining a fourth threshold according to the fact that one half of the standard deviation multiples are used as the standard deviation multiples of the fourth threshold.
Further, the scoring rule is: setting a respective score calculation rule formula for each core index, and obtaining the lowest score when the index result exceeds a first threshold value and a second threshold value of the index;
When the index result is between the first threshold and the third threshold or between the second threshold and the fourth threshold, corresponding scores are obtained by adopting linear calculation according to the index result, the threshold and the score standard, and when the index result is between the third threshold and the fourth threshold, unified standard scores are obtained.
Further, the rank determination module includes:
the relationship obtaining unit is used for obtaining the corresponding relationship between the risk level and the first range value of the financial falsification probability value and the corresponding relationship between the risk level and the second range value of the risk score;
And the level matching unit is used for determining the risk level of the mechanism to be analyzed based on the first matching relation between the current financial falsification probability value and the first range value and the second matching relation between the current risk score and the second range value.
According to the financial data analysis device provided by the embodiment of the invention, the financial data to be analyzed is analyzed by combining the logistic regression model and the Themis financial risk model to obtain the fake probability value and the risk score, the models are jointly applied to obtain the final model result and the risk grade division standard after the two models are jointly applied, the accuracy is effectively improved, the professional and efficient financial data analysis result can be formed, and the risk is found and avoided.
Fig. 5 illustrates a physical schematic diagram of an electronic device, as shown in fig. 5, which may include: processor 510, communication interface (Communications Interface) 520, memory 530, and communication bus 540, wherein processor 510, communication interface 520, memory 530 complete communication with each other through communication bus 540. Processor 510 may invoke logic instructions in memory 530 to perform a financial data analysis method comprising: inputting the financial data to be analyzed of the mechanism to be analyzed into a logistic regression model to obtain a current financial false making probability value; inputting Themis the financial data to be analyzed into a financial risk model to obtain a current risk score; and determining the risk grade of the institution to be analyzed corresponding to the financial data to be analyzed by using the current financial falsification probability and the current risk score.
Further, the logic instructions in the memory 530 described above may be implemented in the form of software functional units and may be stored in a computer-readable storage medium when sold or used as a stand-alone product. Based on this understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution, in the form of a software product stored in a storage medium, comprising several instructions for causing a computer device (which may be a personal computer, a server, a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a usb disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
In another aspect, the present invention also provides a computer program product comprising a computer program stored on a non-transitory computer readable storage medium, the computer program comprising program instructions which, when executed by a computer, enable the computer to perform the method of analysis of financial data provided by the methods described above, the method comprising: inputting the financial data to be analyzed of the mechanism to be analyzed into a logistic regression model to obtain a current financial false making probability value; inputting Themis the financial data to be analyzed into a financial risk model to obtain a current risk score; and determining the risk grade of the institution to be analyzed corresponding to the financial data to be analyzed by using the current financial falsification probability and the current risk score.
In yet another aspect, the present invention also provides a non-transitory computer readable storage medium having stored thereon a computer program which when executed by a processor is implemented to perform the above provided financial data analysis methods, the method comprising: inputting the financial data to be analyzed of the mechanism to be analyzed into a logistic regression model to obtain a current financial false making probability value; inputting Themis the financial data to be analyzed into a financial risk model to obtain a current risk score; and determining the risk grade of the institution to be analyzed corresponding to the financial data to be analyzed by using the current financial falsification probability and the current risk score.
The apparatus embodiments described above are merely illustrative, wherein the elements illustrated as separate elements may or may not be physically separate, and the elements shown as elements may or may not be physical elements, may be located in one place, or may be distributed over a plurality of network elements. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art will understand and implement the present invention without undue burden.
From the above description of the embodiments, it will be apparent to those skilled in the art that the embodiments may be implemented by means of software plus necessary general hardware platforms, or of course may be implemented by means of hardware. Based on this understanding, the foregoing technical solution may be embodied essentially or in a part contributing to the prior art in the form of a software product, which may be stored in a computer readable storage medium, such as ROM/RAM, a magnetic disk, an optical disk, etc., including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the method described in the respective embodiments or some parts of the embodiments.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention, and are not limiting; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims (16)
1. A method of financial data analysis, comprising:
Inputting the financial data to be analyzed of the mechanism to be analyzed into a logistic regression model to obtain a current financial false making probability value;
inputting Themis the financial data to be analyzed into a financial risk model to obtain a current risk score;
determining the risk level of the institution to be analyzed corresponding to the financial data to be analyzed by using the current financial falsification probability and the current risk score;
Inputting Themis the financial data to be analyzed into a financial risk model, and obtaining a current risk score includes:
determining an index result of a core index based on the financial data to be analyzed;
Obtaining a scoring value of the index result based on the index result, the corresponding index threshold and the scoring rule;
adding the scoring values of all indexes to obtain a preliminary risk score of the mechanism to be analyzed;
And adjusting the preliminary risk score by using an adjustment item to obtain the current risk score of the mechanism to be analyzed.
2. The method of claim 1, wherein the step of determining the position of the substrate comprises,
The logistic regression model is obtained by the following steps:
Acquiring sample data; the sample data comprise good samples and bad samples, wherein the good samples are financial and newspaper data of the same industry with the same preset time period of the mechanism to be analyzed; the bad sample is fake financial data of the same industry as the organization to be analyzed;
Preprocessing the sample data;
screening independent variables and dependent variables from the preprocessed sample data;
Determining the correlation coefficient of the independent variable and the dependent variable, and screening out the independent variable with the correlation exceeding a preset value; clustering the independent variables to obtain a new independent variable set;
And performing logistic regression on the new independent variable set and the dependent variable to obtain an effective prediction independent variable set.
3. The method of claim 2, wherein the step of determining the position of the substrate comprises,
The preprocessing of the sample data comprises:
deleting data with the number of data deletions exceeding a preset number in the sample data;
Replacing an abnormal value in the sample data;
And (5) carrying out standardization processing on the sample data.
4. The method of claim 1, wherein the step of determining the position of the substrate comprises,
The core index includes at least one of the following: the balance period level index, the balance period change amount index, the financial debt state level index, the financial debt state change amount index, the physical asset turnover change amount, the investment benefit level, the intangible asset efficiency change amount, the frequent balance ratio, the abnormal value coefficient, the payment remaining force coefficient, the asset coefficient, the cost system and other indexes, the mobile liability occupation ratio and the monetary funds occupation proportion.
5. The method of claim 1, wherein the step of determining the position of the substrate comprises,
The index threshold includes: a first threshold, a second threshold, a third threshold, and a fourth threshold;
The first threshold is a threshold which adopts the first 10% quantile value as an extreme abnormal point for the whole index;
The second threshold is a threshold which adopts the last 10% quantile value as an extreme abnormal point for the whole index;
The third threshold value is determined by the following steps: calculating the mean value of each dimension as a median result after calculating the first 10% of data, calculating standard deviation, respectively calculating standard deviation multiples of a first threshold point, and determining the standard deviation multiples as a third threshold according to the standard deviation multiples which are half of the standard deviation multiples;
the fourth threshold value is determined by the following steps: and calculating the mean value of each dimension as a median result after 10% of data are calculated, calculating standard deviation, respectively calculating standard deviation multiples of a second threshold, and determining a fourth threshold according to the fact that one half of the standard deviation multiples are used as the standard deviation multiples of the fourth threshold.
6. The method of claim 5, wherein the step of determining the position of the probe is performed,
The scoring rule is as follows: setting a respective score calculation rule formula for each core index, and obtaining the lowest score when the index result exceeds a first threshold value and a second threshold value of the index;
When the index result is between the first threshold value and the third threshold value or between the second threshold value and the fourth threshold value, obtaining a corresponding score by adopting linear calculation according to the index result, the threshold value and the score standard;
And when the index result is between the third threshold value and the fourth threshold value, obtaining a unified standard score.
7. The method according to any one of claim 1 to 6, wherein,
The determining the risk level of the institution to be analyzed corresponding to the financial data to be analyzed by using the current financial falsification probability and the current risk score comprises the following steps:
Acquiring a corresponding relation between a risk level and a first range value of a financial falsification probability value and between the risk level and a second range value of a risk score;
And determining the risk level of the institution to be analyzed based on a first matching relation between the current financial falsification probability value and the first range value and a second matching relation between the current risk score and the second range value.
8. A financial data analysis apparatus, comprising:
the probability acquisition module is used for inputting the financial data to be analyzed of the mechanism to be analyzed into the logistic regression model to obtain the current financial false-making probability value;
The score acquisition module is used for inputting Themis the financial data to be analyzed into a financial risk model to obtain a current risk score;
The grade determining module is used for determining the risk grade of the mechanism to be analyzed corresponding to the financial data to be analyzed by utilizing the current financial false making probability and the current risk score;
The score acquisition module includes:
An index determining unit for determining an index result of a core index based on the financial data to be analyzed;
The scoring obtaining unit is used for obtaining a scoring value of the index result based on the index result, the corresponding index threshold value and the scoring rule;
the preliminary obtaining unit is used for adding the scoring values of all indexes to obtain a preliminary risk score of the mechanism to be analyzed;
And the grading adjustment unit is used for adjusting the preliminary risk grading by using the adjustment items to obtain the current risk grading of the mechanism to be analyzed.
9. The apparatus of claim 8, wherein the device comprises a plurality of sensors,
The logistic regression model is obtained by the following steps:
Acquiring sample data; the sample data comprise good samples and bad samples, wherein the good samples are financial and newspaper data of the same industry with the same preset time period of the mechanism to be analyzed; the bad sample is fake financial data of the same industry as the organization to be analyzed;
Preprocessing the sample data;
screening independent variables and dependent variables from the preprocessed sample data;
Determining the correlation coefficient of the independent variable and the dependent variable, and screening out the independent variable with the correlation exceeding a preset value; clustering the independent variables to obtain a new independent variable set;
And performing logistic regression on the new independent variable set and the dependent variable to obtain an effective prediction independent variable set.
10. The apparatus of claim 9, wherein the device comprises a plurality of sensors,
The preprocessing of the sample data comprises:
deleting data with the number of data deletions exceeding a preset number in the sample data;
Replacing an abnormal value in the sample data;
And (5) carrying out standardization processing on the sample data.
11. The apparatus of claim 8, wherein the device comprises a plurality of sensors,
The core index includes at least one of the following: the balance period level index, the balance period change amount index, the financial debt state level index, the financial debt state change amount index, the physical asset turnover change amount, the investment benefit level, the intangible asset efficiency change amount, the frequent balance ratio, the abnormal value coefficient, the payment remaining force coefficient, the asset coefficient, the cost system and other indexes, the mobile liability occupation ratio and the monetary funds occupation proportion.
12. The apparatus of claim 8, wherein the device comprises a plurality of sensors,
The index threshold includes: a first threshold, a second threshold, a third threshold, and a fourth threshold;
The first threshold is a threshold which adopts the first 10% quantile value as an extreme abnormal point for the whole index;
The second threshold is a threshold which adopts the last 10% quantile value as an extreme abnormal point for the whole index;
The third threshold value is determined by the following steps: calculating the mean value of each dimension as a median result after calculating the first 10% of data, calculating standard deviation, respectively calculating standard deviation multiples of a first threshold point, and determining the standard deviation multiples as a third threshold according to the standard deviation multiples which are half of the standard deviation multiples;
the fourth threshold value is determined by the following steps: and calculating the mean value of each dimension as a median result after 10% of data are calculated, calculating standard deviation, respectively calculating standard deviation multiples of a second threshold, and determining a fourth threshold according to the fact that one half of the standard deviation multiples are used as the standard deviation multiples of the fourth threshold.
13. The apparatus of claim 12, wherein the device comprises a plurality of sensors,
The scoring rule is as follows: setting a respective score calculation rule formula for each core index, and obtaining the lowest score when the index result exceeds a first threshold value and a second threshold value of the index;
When the index result is between the first threshold value and the third threshold value or between the second threshold value and the fourth threshold value, obtaining a corresponding score by adopting linear calculation according to the index result, the threshold value and the score standard;
And when the index result is between the third threshold value and the fourth threshold value, obtaining a unified standard score.
14. The device according to any one of claims 8 to 13, wherein,
The rank determination module includes:
the relationship obtaining unit is used for obtaining the corresponding relationship between the risk level and the first range value of the financial falsification probability value and the corresponding relationship between the risk level and the second range value of the risk score;
And the level matching unit is used for determining the risk level of the mechanism to be analyzed based on the first matching relation between the current financial falsification probability value and the first range value and the second matching relation between the current risk score and the second range value.
15. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the steps of the financial data analysis method of any one of claims 1 to 7 when the program is executed by the processor.
16. A non-transitory computer readable storage medium having stored thereon a computer program, which when executed by a processor performs the steps of the financial data analysis method as claimed in any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110164468.4A CN112801529B (en) | 2021-02-05 | 2021-02-05 | Financial data analysis method and device, electronic equipment and medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110164468.4A CN112801529B (en) | 2021-02-05 | 2021-02-05 | Financial data analysis method and device, electronic equipment and medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112801529A CN112801529A (en) | 2021-05-14 |
| CN112801529B true CN112801529B (en) | 2024-05-17 |
Family
ID=75814524
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110164468.4A Active CN112801529B (en) | 2021-02-05 | 2021-02-05 | Financial data analysis method and device, electronic equipment and medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112801529B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113486366A (en) * | 2021-06-08 | 2021-10-08 | 贵州电网有限责任公司 | Web illegal operation behavior detection method based on cluster analysis |
| CN113505823B (en) * | 2021-07-02 | 2023-06-23 | 中国联合网络通信集团有限公司 | Supply chain security analysis method and computer readable storage medium |
| CN114297689B (en) * | 2021-12-29 | 2022-09-20 | 鲁信科技股份有限公司 | Financial wind control method and device based on privacy calculation and storage medium |
| CN116757334B (en) * | 2023-08-16 | 2023-11-24 | 江西科技学院 | Financial data processing method, system, readable storage medium and computer |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2012018968A1 (en) * | 2010-08-06 | 2012-02-09 | The Dun And Bradstreet Corporation | Method and system for quantifying and rating default risk of business enterprises |
| CN108171602A (en) * | 2017-06-28 | 2018-06-15 | 永辉青禾商业保理(重庆)有限公司 | The method for building up and device of accounts receivable factoring business risk model |
| CN109583682A (en) * | 2018-10-11 | 2019-04-05 | 平安科技（深圳）有限公司 | Recognition methods, device and the computer equipment of business finance fraud risk |
| CN111383101A (en) * | 2020-03-25 | 2020-07-07 | 深圳前海微众银行股份有限公司 | Post-loan risk monitoring method, device, equipment and computer-readable storage medium |
| CN111429286A (en) * | 2020-03-20 | 2020-07-17 | 韭韭投研(北京)数据科技有限公司 | Financial reliability evaluation method, device, equipment and storage medium |
| CN112801456A (en) * | 2020-12-30 | 2021-05-14 | 北京知因智慧科技有限公司 | Fund liquidity risk early warning method, device and electronic equipment |
| CN113506173A (en) * | 2021-08-06 | 2021-10-15 | 国网电子商务有限公司 | Credit risk assessment method and related equipment thereof |
| CN114581209A (en) * | 2022-03-02 | 2022-06-03 | 中国建设银行股份有限公司 | Method, device and equipment for training financial analysis model and storage medium |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110099101A1 (en) * | 2009-10-26 | 2011-04-28 | Bank Of America Corporation | Automated validation reporting for risk models |
-
2021
- 2021-02-05 CN CN202110164468.4A patent/CN112801529B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2012018968A1 (en) * | 2010-08-06 | 2012-02-09 | The Dun And Bradstreet Corporation | Method and system for quantifying and rating default risk of business enterprises |
| CN108171602A (en) * | 2017-06-28 | 2018-06-15 | 永辉青禾商业保理(重庆)有限公司 | The method for building up and device of accounts receivable factoring business risk model |
| CN109583682A (en) * | 2018-10-11 | 2019-04-05 | 平安科技（深圳）有限公司 | Recognition methods, device and the computer equipment of business finance fraud risk |
| CN111429286A (en) * | 2020-03-20 | 2020-07-17 | 韭韭投研(北京)数据科技有限公司 | Financial reliability evaluation method, device, equipment and storage medium |
| CN111383101A (en) * | 2020-03-25 | 2020-07-07 | 深圳前海微众银行股份有限公司 | Post-loan risk monitoring method, device, equipment and computer-readable storage medium |
| CN112801456A (en) * | 2020-12-30 | 2021-05-14 | 北京知因智慧科技有限公司 | Fund liquidity risk early warning method, device and electronic equipment |
| CN113506173A (en) * | 2021-08-06 | 2021-10-15 | 国网电子商务有限公司 | Credit risk assessment method and related equipment thereof |
| CN114581209A (en) * | 2022-03-02 | 2022-06-03 | 中国建设银行股份有限公司 | Method, device and equipment for training financial analysis model and storage medium |
Non-Patent Citations (3)
| Title |
|---|
| "基于Themis 模型的财务安全研究-以斯太尔公司为例";程佳琪;《中国优秀硕士学位论文全文数据库 经济与管理科学辑》;正文第18页和第9页 * |
| 创新型中小企业危机预警模型算法改进研究;宋丽平;李美君;;科技与管理;20141115(第06期);全文 * |
| 审计过程中虚假财务报告的回归分析模型研究;舒文存;;西安文理学院学报(自然科学版)(第04期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112801529A (en) | 2021-05-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112801529B (en) | Financial data analysis method and device, electronic equipment and medium | |
| Doumpos et al. | Developing and testing models for replicating credit ratings: A multicriteria approach | |
| CN111476660B (en) | Intelligent wind control system and method based on data analysis | |
| US20080021813A1 (en) | Method for scoring accounts for retention and marketing accounts based on retention and profitability | |
| Callejón et al. | A System of Insolvency Prediction for industrial companies using a financial alternative model with neural networks | |
| Sun et al. | Crude oil price and exchange rate: Evidence from the period before and after the launch of China's crude oil futures | |
| Lee et al. | Predicting the financial crisis by Mahalanobis–Taguchi system–Examples of Taiwan’s electronic sector | |
| Rahman et al. | Identifying financial distress indicators of selected banks in Asia | |
| Bhat et al. | The impact of risk modeling on the market perception of banks’ estimated fair value gains and losses for financial instruments | |
| Lu et al. | Research and improvement of fraud identification model of Chinese A-share listed companies based on M-score | |
| Dimitras et al. | Evaluation of empirical attributes for credit risk forecasting from numerical data | |
| Arlov et al. | Cash flow in predicting financial distress and bankruptcy | |
| Wulandari et al. | The effect of interest rates, exchange rates and capital structure on banking profitability of BUMN and Private Go Public in Indonesia | |
| Elkamhi et al. | A benchmark for collateralized loan obligations | |
| CN113506173A (en) | Credit risk assessment method and related equipment thereof | |
| Mačerinskiene et al. | The evaluation model of a commercial bank loan portfolio | |
| Roca et al. | Risks and challenges of complex financial instruments: an analysis of SSM banks | |
| Zeng | Research on risk measurement and early warning of electronic banking business based on gmdh algorithm | |
| Oladutire et al. | Return on assets and market stock prices of quoted deposit money banks in Nigeria | |
| Kim | Better than Risk-Free: Do Reserve Premiums Crowd Out Bank Lending? | |
| Coelho | Predicting Corporate Failure: an application of Altman's Z-Score and Altman's EMS models to the JSE Alternative Exchange from 2008 to 2012 | |
| Fadjar et al. | Analysis calculation of allowance for impairment losses of credit before and after the implementation of Psak 50 & 55 on profit at bank X | |
| Jackson‐Moore | Measuring operational risk | |
| Mileris | The impact of macroeconomic environment on credit risk in commercial banks | |
| Christiawan et al. | Detecting Earnings Management: A Foreign Exchange Losses (FEL) Model |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |